An Ensemble Learning Scheme for Indoor-Outdoor Classification Based on KPIs of LTE Network

Abstract

Wireless Big Data has aroused extensive attention, as mass mobile devices have been developed and deployed for the upcoming 5G era. The context information of these devices is of importance for personalized services in a smart environment. Nevertheless, the constant change of scenes challenges to the network operator. In this paper, we propose an ensemble learning scheme for indoor-outdoor classification for a typical urban area, based on the cellular data captured in a commercial LTE network. The variables are extracted by network key performance indicators (KPIs) and radio propagation knowledge. Based on these main variables, the decision trees grow and split by the Gini index of sampled features. Then, all decision trees are assembled as weak learners to build the ensemble scheme, thus improving the discrimination ability. The self-validation results show the ensemble model achieves extreme accurate (with an out-of-bag error lower than 1%) classification for indoor and outdoor environments. Moreover, the prominent variables are selected based on the variable importance of in the initial training. The reconfigured model based on fewer variables and less weak learners also gains the highest accuracy and relative short compute time, compared with other classical machine learning methods.INDEX TERMS Propagation measurement, ensemble learning, LTE, channel model, scene classification.

ML-Based Classification of Device Environment Using Wi-Fi and Cellular Signal Measurements

“…For example, Femto cells deployed indoors need to quickly determine what the user-environment is in order to reduce hand-over delay and avoid ping-pong effects [13]- [15]. In [16], an ensemble learning scheme for indoor-outdoor classification is proposed for a specific urban area consisting of five malls, based on the cellular data captured in a commercial LTE network. The variables are extracted by network key performance indicators (KPIs) and radio propagation knowledge.…”

A Novel Approach for User Equipment Indoor/Outdoor Classification in Mobile Networks

“…The solution proposed in [20] uses a single smartphone with the TEMS application [21], resulting in 24912 instances of data, where each instance has: Time stamp, Timing Advance, Latitude and Longitude, Evolved Universal Mobile Telecommunications System Terrestrial Radio Access (E-UTRAN) cell identifier, Reference Signal Received Power (RSRP) and Reference Signal Received Quality (RSRQ). From these data, the authors calculated 20 different features and created a RF of CART trees.…”

“…Although the authors in, [19], [20], [3], [23], use only cell data, there are some technical aspects that this work attempts to improve. The first work from Wang et al [19] considers a dataset that includes only a small geographic area, a university campus, and uses 2G technology, which is phasing-out and has been replaced by both 3G and 4G.…”

“…The first work from Wang et al [19] considers a dataset that includes only a small geographic area, a university campus, and uses 2G technology, which is phasing-out and has been replaced by both 3G and 4G. Regarding, the work of Zhang et al [20], the dataset is more extensive than the previous one, however, it only considers an urban environment of a single city. Additionally, the train and test approach used by the authors, OOB, is debatable and can lead to The novelty presented in [3] depends on the accessibility to an updated map of the city and the ratio's estimation between indoor and outdoor area.…”

Measurement and Simulation for Vehicle-to-Infrastructure Communications at 3.5 GHz for 5G

“…In order to provide the reliable and low latency for the V2X services, efforts have been made in recent years to develop V2X communications using IEEE 802.11p [5] and ITS-G5 [6]. Cellular vehicle-to-everything (C-V2X) communication is now regarded as another promising and feasible solution for the fifth-generation mobile networks-(5G-) enabled vehicular communications [7]. C-V2X will allow vehicles to communicate wirelessly with each other, with traffic signs, and with other roadside infrastructures by using the same 5G networks coming to mobile phones.…”